## Tuning Parameters in Engineering Design (1993)

Venue: | ASME Journal of Mechanical Design |

Citations: | 21 - 9 self |

### BibTeX

@ARTICLE{Otto93tuningparameters,

author = {Kevin N. Otto and Erik K. Antonsson},

title = {Tuning Parameters in Engineering Design},

journal = {ASME Journal of Mechanical Design},

year = {1993},

pages = {1}

}

### Years of Citing Articles

### OpenURL

### Abstract

In the design and manufacture of mechanical devices, there are parameters whose values are determined by the manufacturing process in response to errors introduced in the device’s manufacture or operating environment. Such parameters are termed tuning parameters, and are distinct from design parameters which the designer selects values for as a part of the design process. This paper introduces tuning parameters into the design methods of: optimization, Taguchi’s method, and the method of imprecision [10]. The details of the mathematical formulation, along with a design example, are presented and discussed. Including tuning parameters in the design process can result in designs that are more tolerant of variational noise. 1

### Citations

424 |
Fuzzy sets as a basis for a theory of possibility. Fuzzy Sets and Systems
- Zadeh
- 1978
(Show Context)
Citation Context ...Tuning parameters have a range over which they can vary, and any value within that range can be used, yet the designer does not specify their values. This is by definition a possibilistic uncertainty =-=[12]-=-. Further, degrees of possibility can be introduced. That is, not only is a range of possibility given for tuning parameters, but every value within the range is given a normalized rank indicating how... |

315 |
E.," Multiple criteria optimization: Theory, computation, and application
- Steuer
(Show Context)
Citation Context ... to ensure the variances are of the same order. The relative weighting coefficient used here was √ 5. Determining weighted sum coefficients is incidental to the tuning parameter formulation, refer to =-=[2, 7]-=- for multi-objective function optimization formulations. Substituting Equations (8) and (9) into (1) produces the problem to be solved as: subject to: M ∗ ,K ∗ =min M,K [∫ δk [ |τ − τs| min + x0 τs √ ... |

186 |
Quality Engineering using Robust Design
- Phadke
- 1989
(Show Context)
Citation Context ... design, and perturbations from these nominal values. The method also considers different noise parameter arrangements (NPAs), and selects the DPA which minimizes variance across the NPAs. See [1] or =-=[5]-=-. The basic Taguchi method selects the DPA defined by: [ [ ∑ −10 log (f(DPA, NPA) − τ) 2 ]] (4) DPA ∗ =max DPA NPA for a design in which an objective f must be maintained as close to τ as possible. Wi... |

58 |
Principles of Optimal Design
- Papalambros, Wilde
- 2000
(Show Context)
Citation Context ...e objective of this paper to introduce tuning parameters, and to demonstrate how they can be modeled in various existing formalized methods of engineering design. We will demonstrate how optimization =-=[4]-=-, Taguchi’s method [1], and the method of imprecision [10, 11] can incorporate tuning parameters. 2 Modeling Tuning Parameters In addition to tuning parameters, two other types of parameters are used ... |

53 | Computations with Imprecise Parameters in Engineering Design: Background and Theory
- Wood, Antonsson
- 1989
(Show Context)
Citation Context ...hich the designer selects values for as a part of the design process. This paper introduces tuning parameters into the design methods of: optimization, Taguchi’s method, and the method of imprecision =-=[10]-=-. The details of the mathematical formulation, along with a design example, are presented and discussed. Including tuning parameters in the design process can result in designs that are more tolerant ... |

48 |
Introduction to Quality Engineering, Asian Productivity Organization
- Taguchi
- 1986
(Show Context)
Citation Context ...adio or television signal tuning circuit adjustments, etc. They are characterized by the tuning parameter’s ability to compensate for noise. Note that the term “noise” is used in the sense of Taguchi =-=[8]-=-, meaning that there are three types of noise observed: external, internal, and variational. External noise errors are due to environmental fluctuations, such as operating temperatures, humidities, et... |

39 |
Optimal Engineering Design; Principles and Applications
- Siddall
- 1982
(Show Context)
Citation Context ...e mass M) would be chosen such that the desired performance is achieved as often as possible. This is the extent to which formal techniques such as Taguchi’s method [1] and probabilistic optimization =-=[6]-=- currently model parameter types: select the design parameter values which maximize the output despite the noise. With tuning parameters, however, more freedom is provided to the designer. The above t... |

23 |
The taguchi approach to parameter design
- Byrne, Taguchi
- 1987
(Show Context)
Citation Context ...per to introduce tuning parameters, and to demonstrate how they can be modeled in various existing formalized methods of engineering design. We will demonstrate how optimization [4], Taguchi’s method =-=[1]-=-, and the method of imprecision [10, 11] can incorporate tuning parameters. 2 Modeling Tuning Parameters In addition to tuning parameters, two other types of parameters are used in the design process:... |

20 | Extensions to the taguchi method of product design
- Otto, Antonsson
- 1993
(Show Context)
Citation Context ...on used either P or τ asaconstraint equation (using the aspiration level), and a single objective function optimization performed. Taguchi’s method, of course, cannot incorporate constraint equations =-=[3]-=-. But one additional important question is the tolerance of the device to variational noise. Modeling with tuning parameters allows for more variability in δk,sincex0can be tuned to keep P and τ on ta... |

14 |
Modeling Imprecision and Uncertainty
- Wood, Antonsson
- 1990
(Show Context)
Citation Context ... and to demonstrate how they can be modeled in various existing formalized methods of engineering design. We will demonstrate how optimization [4], Taguchi’s method [1], and the method of imprecision =-=[10, 11]-=- can incorporate tuning parameters. 2 Modeling Tuning Parameters In addition to tuning parameters, two other types of parameters are used in the design process: design parameters and noise parameters.... |

11 |
A Method for Representing and Manipulating Uncertainties in Preliminary Engineering Design
- Wood
- 1989
(Show Context)
Citation Context ... for tuning parameters, but every value within the range is given a normalized rank indicating how possible the value is. The method of imprecision as a preliminary design methodology is presented in =-=[9, 10, 11]-=-. Design parameters values are ranked, by the designer, as to the degree to which they are preferred. These design parameter preferences are then propagated into preferences on multiple performance pa... |

6 |
A.Multi-Criterion Optimization in Engineering with Fortran Examples
- OSYCSKA
- 1984
(Show Context)
Citation Context ... to ensure the variances are of the same order. The relative weighting coefficient used here was √ 5. Determining weighted sum coefficients is incidental to the tuning parameter formulation, refer to =-=[2, 7]-=- for multi-objective function optimization formulations. Substituting Equations (8) and (9) into (1) produces the problem to be solved as: subject to: M ∗ ,K ∗ =min M,K [∫ δk [ |τ − τs| min + x0 τs √ ... |